American scientists from the University of Arkansas discovered that the smart chatbot ChatGPT outperformed humans in creative thinking tests. The study was published in the scientific journal magazine Scientific Reports (SciRep).
The discovery was made during tests designed to demonstrate its ability to produce a unique solution to a question that had no clear answer. An example of such a task might be: “What is the best way to avoid talking about politics with your family?”
151 volunteers participated in the research. They had to confront the GPT-4 neural network model in three tests: suggesting creative uses of objects such as a rope or a fork, explaining possible outcomes of hypothetical situations (for example, what would happen if people no longer needed sleep), and also finding the 10 names that were furthest apart in meaning.
The answers were evaluated according to the number of options, their length and the semantic (meaning) difference between the words.
In the end, the authors found: “Overall, GPT-4 was more original and complex than humans on each of the original thinking tasks. In other words, GPT-4 exhibited greater creativity on all creative thinking tasks.”
Previously ChatGPT fallen He went into delirium that lasted more than 16 hours.
What are you thinking?
Source: Gazeta
Jackson Ruhl is a tech and sci-fi expert, who writes for “Social Bites”. He brings his readers the latest news and developments from the world of technology and science fiction.